Abstract: An integral control system with fuzzy state observer based on Takagi-Sugeno (T-S) fuzzy model was designed in order to improve the load-following capability of nuclear reactor. Linear models were derived from the original nonlinear point reactor kinetics equations on several operating points. The fuzzy integral controller and the fuzzy state observer were designed by using the parallel distributed compensation (PDC) technique, and the T-S fuzzy control system was constructed with the relative neutron density referenced to the equilibrium condition of rated power as the premise variable. The problem that part of state variables can not be measured was successfully solved by introducing fuzzy state observer. The stability analysis was given by means of linear matrix inequality (LMI) approach, and the control system was guaranteed to be stable within the large range. The simulation results demonstrate that the control system works well over a wide region of operation.